The mathematical analysis of plasma HIV-1 viral load decay after the initiation of antiretroviral therapy has led to a number of important insights about the dynamics of HIV-1 infection, but not to a cure. Modeling, coupled to analysis of clinical data, suggested that with 100% effective therapy HIV could be eliminated from both short- and long-lived productively infected cell populations. Here, we propose to examine a set of problems related to the hurdles that have beenencountered to eradicating HIV. We will provide better estimates of drug efficacy and, via modeling,examine the consequences of drug regimes that are not 100% effective, taking into regard potential drug sanctuaries. In many patients plasma viral loads can be driven below 50 copies/ml but transient episodes of viremia ("blips") occur and ultra-sensitive assays detect persistent low levels of virus. We propose to examine the dynamics of viral load changes below 50 copies/ml. We will examine data on the occurrence of blips, extracting information about their amplitude and frequency, and then via modeling evaluate potential underlying causes, e.g., activation of latently infected cells, generation of target cells secondary to infection, and poor drug adherence. Effects of perturbations to the immune system via vaccination or infection will be modeled, with a view toward evaluating the implications for refilling reservoirs. In addition to latently infected cells, viral reservoirs, such as virus adhering to the surfaces of follicular dendritic cells and B cells, may be hindering viral eradication. We propose to calculate the lifetime of such reservoirs and examine means of interfering with the multivalent attachment of HIV to cell surfaces via complement and Fc receptors.